-GenerateRegressionAndSimulationOutput.do: This is a (Stata) do-file that generates RegressionOutput.csv and SimulationOutput.csv. This do-file is a modified version of the do-file AER.do and uses STYL_Final_old.dta, a copy of STYL_Final.dta; see below for more information. Both files, AER.do and STYL_Final.dta, come from Blattman et al. (2017) and were obtained from the webpage of the American Economic Association (https://www.aeaweb.org/articles?id=10.1257/aer.20150503).
-GenerateTable4.m: This (Matlab) script generates Table 4 in the paper using SimulationOutput.csv and SSCI.m.
-GenerateTables3and8.m: This script generates Tables 3 and 8 in the paper using SimulationOutput.csv and SSCI.m; set the variable "table" equal to 3 to obtain Table 3 and equal to 8 (or anything other than 3) to obtain Table 8.
-RegressionOutput.csv: This file stores the coefficient estimates (and the corresponding variance matrix estimate) from the regression of a summary index of antisocial behavior on the treatment variables (and a set of controls) that are also reported in the first row in Panel B of Table 2 in Blattman et al. (2017).  
-SimulationOutput.csv: This file stores the coefficient and corresponding variance matrix estimates from the aforementioned regression over 10,000 bootstrap samples.
-SSCI.m: This Matlab function returns our proposed "short and simple confidence intervals" (SSCIs). For a description of the input arguments, see the function header. This file can also be found on the authors' webpages.
-STYL_Final_old.dta: This is a copy of STYL_Final.dta of Blattman et al. (2017) in .dta-format 115 from Stata 12. (The original STYL_Final.dta is .dta-format 118 from Stata 14, 15, 16, or 17.)
NOTE: All results that appear in the paper (and the online appendix) were obtained using the above code executed with Matlab R2020b and Stata/SE 12.1 on a MacBook Pro running macOS Catalina (Version 10.15.17).